The sand fly, Lutzomyia longipalpis is the vector of Leishmania donovani chagasi, the causative agent of visceral leishmaniasis throughout Latin America. We have found variability among Lu. longipalpis sibling species in the erythema producing quality of the bite. The erythema is produced by a potent vasodilator, maxadilan, present in the saliva. Flies whose bite produces essentially no erythema come from a colony established from flies collected in Costa Rica, where an atypical, non-ulcerating cutaneous form of L. d. chagasi infection has recently been described. 'The parasites isolated from persons with this form of the disease have been found to be identical to those isolated from persons with visceral disease. We hypothesize that the vasodilator, maxadilan has a major effect on the development of visceral leishmaniasis and that flies lacking a form of this peptide with vasodilatory activity are less effective vectors of the visceral form of the disease. We propose to study the distribution of genetic variability in Lu. longipalpis in nature and to explore the role of the vasodilator, maxadilan in the development of visceral leishmaniasis. The distribution of sibling species and the recognition and taxonomic status of heretofore unrecognized variants will be studied by surveying isozyme variability at sites in Mexico, Costa Rica, Honduras and Colombia. Species status of distinct forms will be established by the distribution of isozyme genotypes and by cross-mating studies. Vasodilatory activity of saliva among study populations will be determined by injecting salivary gland lysates into the shaved back of white laboratory rabbits and comparing the resulting erythema to standards. The relative amount of maxadilan in saliva of wild caught flies will be determined by Western blot using antimaxadilan antibody probes. Polymorphisms in the maxadilan gene will be determined by studying nucleotide diversity among study populations, by direct DNA sequencing of PCR amplified maxadilan DNA using a cycle-sequencing procedure. In addition we will assay nucleotide diversity by conducting analysis of single DNA strand conformation polymorphisms (SSCP). Specific maxadilan alleles will be correlated with the vasoactivity of saliva by conducting bioassay and PCR amplification of the maxadilan gene from individual flies. The effect of Lu. longipalpis saliva on the development of experimental L. d. chagasi infections in hamsters and BALB/c mice will be studied. Infections will be established both by bite and by injection of parasites mixed with appropriate dilutions of salivary gland lysates.